Fast Trade UX Research & Evaluation

As part of Bitkub’s application usability testing initiative, I participated in pre- and post-launch UX research focused on evaluating key trading experiences across the platform.

The overall testing scope included:

  • Fast Trade

  • Trade

  • Deposit / Withdraw THB

  • Deposit / Withdraw Crypto

  • Basic vs advanced layout

Overview

My primary focus was evaluating the Fast Trade experience through moderated usability testing, behavioral observation, insight analysis, and emotional journey mapping. The feature had already been designed before I joined, so my role was to understand how users discovered, interpreted, and interacted with it, then translate those findings into UX opportunities and recommendations.

My Main Repsonsibilities

Objectives

The goal of the research was to:

  • Understand how users interact with Fast Trade

  • Identify confusion points during discovery and navigation

  • Evaluate whether users understand the feature’s purpose

  • Uncover opportunities to improve accessibility and clarity

Research Process

Usability Testing

We conducted task-based usability sessions where participants completed realistic trading-related scenarios using the Bitkub application. This was tested with 71 users.

Example task:

“Purchase ETH worth 5,000 THB using Fast Trade.”

During sessions, we observed:

  • Navigation behavior

  • Hesitation points

  • Misclicks

  • Confusion moments

  • Completion confidence

SUS Survey

After each session, participants completed a SUS questionnaire to evaluate:

  • Perceived usability

  • Ease of learning

  • Confidence while using the app

  • Overall satisfaction

This helped us combine qualitative observations with quantitative usability feedback.

Persona Creation

Based on findings from the overall testing initiative, we created personas representing different trading behaviors and experience levels, including:

  • Beginner investors

  • Long-term holders (HODL)

  • Swing traders

  • Active traders

Although the personas were not created specifically for Fast Trade, one consistent pattern appeared across nearly all user groups:

  • Users struggled to find Fast Trade quickly

  • Users were unclear about what the feature actually was

  • Users often could not differentiate it from standard Trade

This became an important insight because the problem occurred regardless of trading experience level.

Key Insight

Through obersavtion and utilising Maze; research revealed that the main issue was not usability inside the feature, but discoverability before entering it.

Across different persona types:

  • Users had difficulty locating Fast Trade

  • Users misunderstood its purpose

  • Users were unsure when they should use Fast Trade instead of standard trading

However, once users entered the flow:

  • Most participants completed transactions successfully

  • Users described the experience as simple and straightforward

This indicated that:

  • The interaction flow itself was relatively easy to use

  • But the surrounding navigation and feature positioning created friction beforehand

Customer Journey Maps

To visualize user behavior and emotional changes throughout the experience, I created customer journey maps based on testing observations.

The journey maps highlighted:

  • Frustration while searching for Fast Trade

  • Uncertainty around terminology and feature differences

  • Increased confidence after entering the flow

  • Lingering confusion about when Fast Trade should be used

These artifacts helped stakeholders quickly understand where friction occurred during the user journey.

Recommendation & Trade-off

Maze analytics showed a 59.3% misclick rate, suggesting the issue was not the Fast Trade flow itself, but how users discovered and understood it.

I recommended treating this as an information architecture problem by revisiting homepage hierarchy and how Fast Trade is positioned against standard Trade.

Due to launch timelines, the team shipped a lighter solution with a visual badge. At post-test launch, click-through improved from 12% to 21%, but I viewed this as an interim fix rather than a full solution to the underlying mental model issue.